CatalogWantedMirrorsLedgerOperating StatementDocsStudio statusCard system
alt.freeweights.text / ColGrep-Agent-2B-MLX-8bit
From: mlx-community <huggingface.co/mlx-community>Newsgroups: alt.freeweights.textSubject:

ColGrep-Agent-2B-MLX-8bit

Date: 2026-11-15Commit: cb6c7b6Signed: 2DC28CC7B26A283B ✓

Text generation model with 2.5B parameters. Safetensors weights, 2.68 GB in all. Published by mlx-community under Apache-2.0.

2.5B params2.68GBColGrep-Agent-2B-MLX-8bitsafetensorspermissive
FactsParams2.5BSize2.68 GBFormatssafetensorsTasktext generationLicenseApache-2.0From sourcepermissive
$ fw get mlx-community/ColGrep-Agent-2B-MLX-8bit
# manifest sha256 094be13ed985… · 2.68 GB · 10 files
$ fw verify ./ColGrep-Agent-2B-MLX-8bit mlx-community/ColGrep-Agent-2B-MLX-8bit
From sourceHosted (signed)mlx-community/ColGrep-Agent-2B-MLX-8bitLast check: 2026-11-15 05:19 UTC
Run at homeGet help

Get help: custom fine-tune or private deploy. We review your setup before we quote.

Quantized of colgrep-agent-2B.

Works with

Run24Fine-tune29Convert15Evaluate13Scan13Apps58
OverviewModel cardFilesUse itLicenseProvenanceModel tree

Overview

Size
2.68 GB
Weights
2.67 GB (safetensors)
Parameters
2.5B
Formats
safetensors
Task
text generation
License
Apache-2.0 (permissive)
Files
10, each with a sha256
Publisher
mlx-community
Added
2026-11-15
Section
alt.freeweights.text
Lineage

Original release from the publisher.

Base model: lightonai/colgrep-agent-2B

Where these bytes come from
Publisher at the pinned commit→You

Sources are tried in order. A file counts only when its size and sha256 match the manifest. Sources are checked when you ask for them, not on a timer.

Sources
BadgeHostLast checkResult
From sourcehuggingface.co2026-11-15 05:19 UTCpassed

Model card

From the publisher at commit cb6c7b6. Shown as plain text with links; images open at the publisher.

ColGrep-Agent-2B - MLX 8BIT#

Native Apple Silicon MLX conversion of LightOn's ColGrep-Agent-2B, a specialized repository code-localization agent built on MiniCPM5-2B.

  • Upstream checkpoint: lightonai/colgrep-agent-2B at f798a61129c6decb5166864e0a3c98373e62c764
  • Conversion: MLX-LM 0.32.0, 8-bit affine, group size 64
  • Compatibility: source chat_template.jinja preserved; top-level rope_theta=5000000 added so MLX-LM's Llama RoPE implementation matches the source configuration.
  • Original model authors: LightOn AI; original model license: Apache-2.0.

Usage#

pip install 'mlx-lm==0.32.0' 'transformers>=5.6.2,<6'
mlx_lm.generate --model . --prompt 'Hello' --max-tokens 128

This is not a general coding assistant. It was trained for iterative code search with colgrep, terminal, and finish. Use the upstream chat template, system prompt, tool schemas, and thinking disabled. The official colgrep --agent CLI currently runs GGUF through llama.cpp; this MLX checkpoint is not a drop-in replacement for that CLI. An MLX-backed harness is separate integration work.

This checkpoint is not claimed to match upstream BF16 on benchmarks until an apples-to-apples evaluation of full search trajectories is reported. The smoke test validates basic load, format, and first tool-call structure only.

Files

10 files · 2.68 GBWeights 2.67 GB (safetensors). Sizes and sha256 from the signed manifest.
Hide all 10 files with their sha256
FileSizesha256
.gitattributes1.52 KB11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361
CONVERSION_RECEIPT.json692 B7ed06e5c9510705b60ed0ed0ffabd833dbf3e2f820a732657815ecb8db711ccc
README.md1.73 KB28b4fd247ace0206b489da8535631303d9e0c64429954f2f555bb590b4df8444
chat_template.jinja9.06 KBcc945752db555d60949b16989df4ccfeb52a313d6b4b5c5229dd786e2e9fcf1c
config.json989 Bd8e805bf384acdc7d4042ef9d78472743aff40da3e59d3e5c7ebeb67c1f03036
generation_config.json237 B8643a6f784ffacefd3441fe1225237ea2123f107709be2bbcb36bf461892a042
model.safetensors2.67 GBd24593d6d59596260e9df3383525f5726930651789a0204094c82e5e42fc8eb2
model.safetensors.index.json68.7 KB50747e37090508f851cdea2f4d5a71c59c2a26a09c8e95c505371ac669e2ac36
tokenizer.json9.89 MB3e065a558a034185fe299917b398685c1facd0169a9eea1e629eb30c171fed81
tokenizer_config.json435 Bb89503c3e5070c6b6d33daf2e20cb4a5c88537c1670d9b7e0cfb4506a61448a9
VersionsEach time the pinned upstream commit or file list changed in this repo. Newest first.
Upstream commitDateContents
cb6c7b6ceb pinned now2026-11-15 05:19 UTC10 files · 2.68 GB

Use it

$ fw get mlx-community/ColGrep-Agent-2B-MLX-8bit
# manifest sha256 094be13ed985… · 2.68 GB · 10 files
$ fw verify ./ColGrep-Agent-2B-MLX-8bit mlx-community/ColGrep-Agent-2B-MLX-8bit
fw.pyChecks the signature and every sha256, tries each source in order, resumes.
$ python3 fw.py get https://freeweights.allternit.com/manifests/mlx-community/ColGrep-Agent-2B-MLX-8bit.json --dir ./ColGrep-Agent-2B-MLX-8bit $ python3 fw.py verify ./ColGrep-Agent-2B-MLX-8bit ./mlx-community__ColGrep-Agent-2B-MLX-8bit.json
hfThe Hugging Face CLI with one setting. Same repo id.
$ HF_ENDPOINT=https://freeweights.allternit.com hf download mlx-community/ColGrep-Agent-2B-MLX-8bit --revision cb6c7b6ceba55b0203b3522efa28e8bd6b1dccbd --local-dir ./ColGrep-Agent-2B-MLX-8bit
transformersLoad the weights in Python.
import os os.environ["HF_ENDPOINT"] = "https://freeweights.allternit.com" from transformers import AutoModelForCausalLM, AutoTokenizer tok = AutoTokenizer.from_pretrained("mlx-community/ColGrep-Agent-2B-MLX-8bit", revision="cb6c7b6ceba55b0203b3522efa28e8bd6b1dccbd") model = AutoModelForCausalLM.from_pretrained("mlx-community/ColGrep-Agent-2B-MLX-8bit", revision="cb6c7b6ceba55b0203b3522efa28e8bd6b1dccbd", device_map="auto")
vLLMServe an OpenAI-compatible API.
$ HF_ENDPOINT=https://freeweights.allternit.com vllm serve mlx-community/ColGrep-Agent-2B-MLX-8bit --revision cb6c7b6ceba55b0203b3522efa28e8bd6b1dccbd
Run at home

License

Name
Apache License 2.0
Id
Apache-2.0
Class
Permissive
Redistribution
yes
Listing
Hosted (signed)
Decided
License as declared by the publisher. No person read it. 2026-11-15
Text in
README.md
Text sha256
28b4fd247ace0206b489da8535631303d9e0c64429954f2f555bb590b4df8444
Terms gate
No extra terms beyond the license.

Read the license text

Licenses are checked against the license table. A custom license passes only after a person reads it.

Provenance

From: mlx-community <huggingface.co/mlx-community>Newsgroups: alt.freeweights.textSubject: ColGrep-Agent-2B-MLX-8bitDate: 2026-11-15Commit: cb6c7b6Signed: 2DC28CC7B26A283B ✓
Upstream
mlx-community/ColGrep-Agent-2B-MLX-8bit
Pinned commit
cb6c7b6ceba55b0203b3522efa28e8bd6b1dccbd
Manifest
ColGrep-Agent-2B-MLX-8bit.json · .minisig
Manifest sha256
094be13ed985e9b8df0b72b1d7cff50f725a2e7484b42553d9155eb2569ec98c
Signature
minisign, Ed25519, key 2DC28CC7B26A283B
Created
2026-11-15 05:19 UTC
Generator
freeweights-manifest/0.1.0
Takedowns
None.

How to verify

$ curl -O https://freeweights.allternit.com/manifests/mlx-community/ColGrep-Agent-2B-MLX-8bit.json $ curl -O https://freeweights.allternit.com/manifests/mlx-community/ColGrep-Agent-2B-MLX-8bit.json.minisig $ minisign -Vm ColGrep-Agent-2B-MLX-8bit.json -P RWQ7KGqyx4zCLXtUFOqvfv8L3g+8fpKJ1FsuxPyOYMF+4KufbUxKq1Oq $ shasum -a 256 <file> # compare with the Files tab

A manifest whose signature fails must be rejected. A file whose size or sha256 does not match must be discarded. Verify a download

Model tree

Parent

lightonai/colgrep-agent-2B · this model is a quantization of it

No model in the catalog names this one as its parent.

Get it
Download
$ HF_ENDPOINT=https://freeweights.allternit.com hf download mlx-community/ColGrep-Agent-2B-MLX-8bit
Open in Freeweights Studio

Pull, verify and run it in the Allternit app on Desktop, web or phone.

Signed manifest
minisign, key 2DC28CC7B26A283B. ColGrep-Agent-2B-MLX-8bit.json. Provenance. Takedown history: none.
Source
huggingface.co/mlx-community/ColGrep-Agent-2B-MLX-8bit

Related models

All of alt.freeweights.text →
From: laion <huggingface.co/laion>Subject: a2-rl-e2egit_large-50-32B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_e2egit-large-15-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_e2egit-v2-10-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_pymethods2test-large-80-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_exp_rpt_unitsyn-python-large-10-8B
From: laion <huggingface.co/laion>Subject: a3-rl-DCAgent_mix_h2_language_proportional-65-8B